# Langfuse

> Open-source tracing, evaluation and prompt management for LLM apps and agents

Langfuse is an observability and evaluation platform for LLM applications and AI agents. It records traces of model calls and tracks cost and latency. It also manages versioned prompts and runs evaluations. It is available as a hosted cloud or self-hosted with Docker, Kubernetes or Terraform.

## Features
- Tracing of LLM calls and agent runs
- Versioned prompt and skill management
- Evaluations including LLM-as-a-judge
- Cost, latency and quality dashboards
- AI Gateway for model provider calls (beta)

## Facts
- Website: https://langfuse.com
- Made by: Langfuse (ClickHouse, Inc.)
- Category: AI agents & assistants
- Pricing: Freemium (from $29/mo)
- Runs on: Web, Self-hosted, API / CLI
- Licence: MIT
- GitHub: https://github.com/langfuse/langfuse (37219 stars)
- Alternative to: LangSmith
- Stage: Live

Listing page: https://launchwoo.com/p/langfuse
Listed on LaunchWoo, a directory of AI apps, developer tools and AI devices: https://launchwoo.com/launches